1. Field of the Invention
This invention relates to an automatic focus adjusting device.
2. Description of the Related Art
The advancement of technology related to various video apparatuses including video cameras, electronic cameras, etc., have been remarkable during recent years. As a result, it has become a standard practice to provide these apparatus with an automatic focus adjusting (AF) function for improvement in performance and operability.
According to a focusing method most popularly employed for the automatic focus adjusting (focusing) devices of these apparatuses, the sharpness of a picture is detected from a video signal obtained through photo-electric conversion by an image sensor or the like and then the position of a focusing lens is controlled and adjusted in such a way as to make the detected sharpness into a maximum degree.
The degree of sharpness is represented by a sharpness signal. The sharpness signal is evaluated in general either by detecting the intensity of a high frequency component extracted from a video signal through a band-pass filter thereinafter referred to as BPF) or by detecting the width of blur of the video signal (width of the edge part of an object image) through a differentiation circuit or the like.
In shooting an ordinary object of shooting, if the lens is out of focus, the level of the high frequency component of the video signal is low and the width of blur is wide. The level of the high frequency component increases and the width of blur decreases accordingly as the lens comes closer to an in-focus state. The level and the width respectively reach a maximum value and a minimum value when the lens reaches a completely in-focus state.
The focusing lens is controlled to drive it at a highest possible speed in the direction of increasing the sharpness if the sharpness is low and the driving speed is lowered accordingly as the sharpness increases in such a way as to bring the focusing lens to a stop just on the top of a hill-like curve representing the degrees of sharpness. This method of control is generally called a hill climbing automatic focusing method or a hill climbing AF method. The adoption of an automatic focusing device operating in accordance with this method has greatly enhanced the operability of an apparatus for taking moving images such as a video camera. This automatic focusing function has become indispensable to a video camera or the like.
FIG. 1 is a block diagram showing one example of the conventional automatic focus adjusting (focusing) device of the kind arranged to use a video signal in a video camera. The illustration includes lens groups 101, 102, 104 and 105. The lens groups 101 and 104 are fixed lens groups. The lens group 102 is arranged to perform a magnifying power varying action (hereinafter referred to as a variator lens). Reference numeral 103 denotes a diaphragm. The lens group 105 is arranged to perform a focus adjusting action and to compensate a focal plane obtained after the magnifying power varying action (hereinafter referred to as a focusing lens).
Reference numeral 106 denotes an image sensor. An AF control microcomputer 107 is arranged to control a whole system including lens driving control, AF control, etc. An aperture encoder 108 is arranged to detect the aperture value of the diaphragm 103. An amplifier 109 is arranged to amplify a signal outputted from the aperture encoder. A conversion circuit 110 is arranged to convert the signal of the aperture encoder 109 into a DC signal of a variable level. An A/D (analog-to-digital) converter 128 is arranged to A/D convert the output of the conversion circuit 110 into a digital signal and to supply the digital signal to the AF control microcomputer 107.
A buffer amplifier 111 is arranged to amplify or impedance-convert the output of the image sensor 106. A BPF 113 is arranged to take out a high frequency component of a video signal which is outputted from the image sensor 106, the high frequency component being arranged to be used for AF control. An AF signal processing circuit 114 is arranged to form from the high frequency component a sharpness signal which is to be used in carrying out the AF control. An A/D converter 115 is arranged to A/D convert the output of the AF signal processing circuit 114 into a digital signal and to supply the digital signal to the AF control microcomputer 107.
A variator lens driving motor 119 and a focusing lens driving motor 122 are respectively arranged to drive the variator lens 102 and the focusing lens 105. Racks 120 and 123 are connected to the variator lens 102 and the focusing lens 105, respectively, and constantly mesh with the rotation shafts of the variator lens driving motor 119 and the focusing lens driving motor 122. Drivers 121 and 124 are arranged to drive the variator lens driving motor 119 and the focusing lens driving motor 122 in accordance with instructions received from the AF control microcomputer 107. An integrator 125 is arranged to integrate a signal outputted from an AGC (automatic gain control) circuit 112. A diaphragm control circuit 126 is arranged to control the aperture of the diaphragm 103 in such a way as to give an adequate amount of exposure in reference to a signal outputted from the integrator 125. A driver 127 is arranged to drive the diaphragm 103. The output signal of the A/D converter 115 to be inputted to the AF control microcomputer 107 has a value which varies with the magnitude of the high frequency component of the video signal. The amount of the high frequency component becomes a maximum amount when the lens is perfectly in focus and becomes smaller when it is out of focus.
In the arrangement described above, the output signal of the A/D converter 115 is called a focus voltage or a focus signal. The AF control microcomputer 107 is arranged to cause the focusing lens 105 to be moved in such a way as to make the value of the output signal of the A/D converter 115 (the focus signal) to become a maximum value. Further, depending on the state of a zoom switch which is not shown, the AF control microcomputer 107 outputs and gives a driving instruction to the drivers 121 and 124 to move the variator lens 102 toward its telephoto end position or toward its wide-angle end position.
A focusing action is performed in the following manner. In the case of an automatic focusing device of the kind arranged to monitor the increase or decrease of the amount of the high frequency component (focus voltage) as in the case of the arrangement shown in FIG. 1, the focusing lens is moved to cause the amount of the high frequency component to become a maximum amount as mentioned above. The amount of the high frequency component increases or decreases in relation to the position of the focusing lens, for example, as represented by a curve 201 in FIG. 2. FIG. 7 is a flow chart showing in outline a flow of processes of the automatic focusing action. Referring to FIG. 7, the focusing action is described as follows. Assuming that the focusing lens has been in repose with an in-focus state obtained for an object of shooting, in cases where the object changes and the focusing lens is moved by driving it again to maximize the focus voltage thereinafter, this process will be called "restarting the focusing lens"), the focusing action must be performed through the following processes.
(I) A check is made to find if the current position of the focusing lens deviates from an in-focus position (a step 706 of FIG. 7).
(II) If so, a check is made to find whether a position where the focus voltage becomes a maximum value is located closer to a nearest distance position or closer to an infinity distance position than the current position of the focusing lens (a step 701 of FIG. 7).
(III) The lens is moved toward the in-focus position in the hill-climbing manner and is brought to a stop at a point where the focus voltage comes to show its maximum value (steps 702 to 705 of FIG. 7).
The details of the process (I) are as follows.
Referring to the hill-like curve 201 of the focus voltage shown in FIG. 2, the hill of the focus voltage changes as represented by a curve 202 when the object of shooting moves. The amount of the focus signal obtained at the focusing lens position then changes by a value A as shown at a part 203 in FIG. 2.
A threshold level 204 is set for making a decision as to whether the focusing lens is to be restarted or not, as represented by a line 204 in FIG. 2. This threshold level 204 is determined according to a focus signal obtained when an in-focus state is last obtained. Assuming that the level of the last focus signal is X, the threshold level 204 can be expressed by the following formula.
A=X-X.times.N/100 (1)
In the formula (1) above, "N" represents a constant predetermined according to the position of the focusing lens and that of the variator lens. The larger the value of the constant N, the more hard the lens is to move.
FIG. 5 shows a relationship obtained among the object distance, the focusing and variator lens in a rear-focus type lens system having the focusing lens located rearwardly of the variator lens. The position of the variator lens (focal length) is shown on the axis of abscissa. The position of the focusing lens is shown on the axis of ordinate. As apparent from FIG. 5, in a case where the lens position is on the wide-angle side and the object of shooting is located in the neighborhood of an infinity distance, the amount of change taking place in the position of the focusing lens for a change taking place in the object distance is so small that an in-focus state might be attained without moving the focusing lens. The value of the constant N is, therefore, set at a larger value for positions on the wide-angle side and at a smaller value for positions on the telephoto side.
When the value of the focus signal changes to a great extent from the threshold level set in the above-stated manner, as indicated by a curve 205 in FIG. 2, it is decided to be necessary to restart the focusing lens, and the direction of driving is selected as described below.
Further, in deciding the restart, it is assumed that a panning operation or the like is quickly performed on the camera in a case where the focus voltage becomes much lower than the threshold level, as indicated by a curve 206 in FIG. 2. In such a case, the blurring time of the object image is minimized for adequate and smooth focusing by not making any check for an in-focus state during the process of selecting the direction of the restart.
Next, the process (II) of selecting the focusing lens driving direction is described as follows. FIG. 3 shows the position of the focusing lens in relation to changes taking place in the level of the focus signal.
Referring to FIG. 3, in a case where the focusing lens is located in a position 308 which is closer to an infinity distance position than an in-focus position, the focusing lens is moved along a locus 305 defined by points (1) to (5). This moving action is called wobbling. Then, since the focusing lens comes nearer and farther to and from the in-focus position, the focus voltage varies as represented by a curve 306 in FIG. 3. If the focusing lens is located in a position closer to the nearest distance position than the in-focus position, on the other hand, the wobbling action performed in the same manner as represented by the locus 305 causes the focus voltage to vary as represented by a curve 307. Comparison of the curves 306 and 307 shows a difference of 180 degrees in phase between the focus-voltage increasing and decreasing curves 306 and 307. In other words, in selecting the driving direction, the wobbling action is carried out in a predetermined manner and a discrimination can be made between a near-focus state and a far-focus state by making a check to find how the focus voltage comes to vary.
Further, a check can be made for an in-focus state by carrying out the wobbling action.
In FIG. 3, a curve 309 shows changes taking place in the focus voltage when the wobbling action is performed at an in-focus point. At the in-focus point, the focus signal shows inphase changes in response to deflections in either of the different directions. When the focus voltage is detected to vary as represented by the curve 309, therefore, the lens is decided to be in an in-focus state and is brought to a stop without performing the process of detecting the maximum value of the focus voltage in the manner as described below.
Next, the process (III) of detecting a point at which the focus voltage is at its maximum value is described as follows. The maximum value of the focus voltage greatly fluctuates according to the object of shooting or the conditions of shooting. Therefore, it is impossible to consider a value of the focus voltage to be the maximum value when any value of the focus voltage is obtained. Therefore, as shown in FIG. 4, while the focusing lens in on the move in the direction selected by the wobbling action, the focus voltage is constantly peak-held. The moving direction of the focusing lens is then reversed at a point of time when the focus voltage comes to change from an increasing state to a decreasing state. The lens is thus returned until the focus voltage becomes equal to the peak-held value before stopping it.
The automatic focusing (focus adjustment) is performed by carrying out control in this manner.
In some cases, while the focusing lens is in an in-focus state, a slight movement of the object of shooting comes to lower the focus voltage to such a small extent that cannot be determined to indicate a defocused state. In such a case, according to a known method, the focusing lens is slowly moved to a minute extent from its current position to reconfirm the in-focus state. This method makes it possible to reliably maintain an in-focus state even for a slow change taking place in the object.
Products on which zoom lenses of high magnifying power, such as 10 magnifications or 12 magnifications, are mounted thereon have increased in number. In the case of such a high magnifying power, shooting with the camera held by hand tends to result in an image shake. The image shake lowers the focus voltage and might accidentally cause a restart of the driving action on the focusing lens. In view of this problem, such cameras are generally provided with an image stabilizing mechanism.
However, the conventional device of the kind described above has had the following shortcomings or problems in carrying out a series of processes.
(i) When the focus signal is caused to vary by wobbling, a check is made to judge whether the top of the hill of the focus signal is located on the side of the nearest distance position or on the side of the infinity distance position. While the judgment can be accurately made in the neighborhood of an in-focus position, the focusing direction tends to be misjudged in the event of an extremely defocused state in which the focusing lens is located within a skirt area of the hill, for example, as shown in FIG. 6. In such a case, changes in the focus signal resulting from wobbling do not readily appear and the focusing direction is apt to be misjudged under the influence of a noise or the like. If the result of the wobbling action happens to be in error, the hill climbing control is executed by climbing up to the top of the hill of the focus signal in the wrong direction in search of an in-focus point. The search in the wrong direction makes the focusing time longer and might bring the focusing lens to a stop while the lens is still in a defocused state.
(ii) While a panning operation is in process, the object of shooting is virtually in a moving state. With panning performed at an adequate speed, the degree of fluctuations taking place in a video signal variously changes. If the wobbling action is performed under such a condition, it is hardly possible to accurately determine the change of the focus signal to be a real change or a spurious change resulting from the moving state of the object of shooting. In such a case, a misjudgment is apt to be made due to a noise.
(iii) In detecting a quick panning operation mentioned above and in carrying out control after the detection, if the quick panning operation is performed for two different objects of shooting located at the same distance from the camera, for example, a drop in the focus voltage taking place during the process of panning would cause the focusing lens to be accidentally driven even when the camera is brought to a stop at the end of the panning in a state of being directed to one of the objects for which the focusing lens does not have to be moved. Such panning gives an unstable impression.
(iv) In the case of the rear-focus type lens (FIG. 5), if the wobbling action is performed for an object of shooting located at such a part for which the position of the focusing lens is on the side of the infinity distance position with the variator lens in a wide-angle position, a relation between the number of focusing lens driving pulses and the object distance inevitably causes the wobbling action to show up too much when deflection is effected to an amplitude by which the focus adjusting direction can be decided.